Multi-Branch Network with Ensemble Learning for Text Removal in the Wild

Multi-Branch Network with Ensemble Learning for Text Removal in the Wild

연구 분야: Strategies

학회: Asian Conference on Computer Vision

The scene text removal (STR) is a task to substitute text regions with visually realistic backgrounds. Due to the diversity of scene text and the intricacy of background, earlier STR approaches may not successfully remove scene text. We discovered that different networks produce different text removal results. Thus, we present a novel STR approach with a multi-branch network to entirely erase the text while maintaining the integrity of the backgrounds. The main branch preserves high-resolution texture information, while two sub-branches learn multi-scale semantic features. The complementary erasure networks are integrated with two ensemble learning fusion mechanisms: a feature-level fusion and an image-level fusion. Additionally, we propose a patch attention module to perceive text location and generate text attention features. Our method outperforms state-of-the-art approaches on both real-world and synthetic datasets, improving PSNR by 1.78 dB in the SCUT-EnsText dataset and 4.45 dB in the SCUT-Syn dataset.